Glean Revenue and Valuation: How Glean Grew to $300M ARR

Glean grew by solving a simple problem that became harder as companies adopted more software: employees could not find the information they needed.

The company started with enterprise search in 2019. It connected tools such as Google Drive, Slack, Salesforce, Jira, and GitHub, then let employees search them from one place.

That alone was hard to build. Glean had to understand permissions, rank information correctly, and learn how people and documents inside each company were related.

For its first few years, Glean worked closely with a small group of design partners and sold mainly to fast-growing technology companies.

Then ChatGPT launched.

Suddenly, the search infrastructure Glean had spent years building could do much more than return documents. It could give employees direct answers, generate content, and eventually run AI agents across company systems.

Growth accelerated.

Glean reported reaching $100 million in annual recurring revenue in early 2025, $200 million later that year, and $300 million in May 2026.

Its latest disclosed private-market valuation is $7.2 billion.

The important part of the story is what happened before the AI boom.

Glean had already spent years building the data connections, permissions system, search index, enterprise graph, customer relationships, and sales motion that made its AI products useful.

Glean's growth at a glance

The progression is easier to understand as five stages:

enterprise search → free design partners → CIO-led expansion → AI assistant → AI agents

Each stage increased the value of the infrastructure Glean had already built.

The idea for Glean came from a problem inside Rubrik

Before Glean, Arvind Jain helped start Rubrik.

Jain had spent more than a decade at Google, where he worked on Search, Maps, and YouTube. At Rubrik, he led research and development.

Rubrik grew quickly.

As the company added employees, Jain noticed something strange: headcount was increasing faster than productivity.

Rubrik surveyed employees to understand why.

One of their biggest problems was finding information.

The company used hundreds of cloud applications. A document might be in Google Drive. A conversation might be in Slack. Product information could be in Jira. Customer data might live in Salesforce.

Employees first had to guess where the information was stored. Then they had to search each system separately.

Jain understood the problem because of his background in search.

He went looking for software that could search across Rubrik's tools.

He could not find anything good enough.

"I actually began my journey as a prospective Glean customer, except there was no Glean product available to purchase," he later wrote.

So Jain started Glean in 2019 with T.R. Vishwanath, Piyush Prahladka, and Tony Gentilcore.

The initial idea was easy to explain:

Google search for your company.

That simple positioning would become important.

Glean started with a narrow customer segment

Glean did not begin by trying to sell to every company.

Its first target was fast-growing technology companies.

Jain said the early sweet spot was usually a company with around 500 to 2,000 employees.

These companies had enough employees, applications, and documents for information sprawl to become painful.

They were also young enough to try software from another startup.

Glean understood these users well. Its founders had worked inside similar technology companies and knew what engineers, salespeople, support teams, and marketers searched for every day.

That made the initial customer profile unusually clear.

A 20-person startup probably did not need an enterprise-wide search engine.

A giant bank might need one badly, but selling to it would require long security reviews, procurement cycles, and integrations.

Fast-growing technology companies sat in the middle.

They had the pain and could still move quickly.

Glean gave its first customers the product for free

The founders still had a problem.

Enterprise search was not a hot software category.

Companies knew finding information was frustrating. They did not necessarily have a budget called "enterprise search."

Glean solved this by removing the buying decision at the beginning.

Jain said the company worked with about 20 design partners during its first two years and gave them Glean for free.

These were not self-serve users signing up for a free plan.

Glean worked directly with each company, connected its systems, watched how employees used the product, and improved it.

The goal was to answer a more important question than whether someone would sit through a sales call:

Would employees keep using Glean once it was installed?

The answer started to look promising.

An early Glean employee recalled that onboarding the first pilot customers required following a long set of commands from a Google Doc. If everything worked, it could take a full day.

The team kept reducing that work.

At some early design partners, employees became so dependent on Glean that Jain later recalled users objecting when security teams considered turning it off.

The story is based on Jain's retrospective account, so it should not be treated as independent evidence. But it captures the behavior Glean was looking for.

People were building a habit around search.

That gave Glean something much more useful than a list of interested prospects: active users and customer references.

Free pilots turned into paid enterprise contracts

Glean started converting its design partners into paying customers around 2021.

Early customers included companies such as Confluent, Okta, and Grammarly.

By the time Glean came out of stealth in September 2021, more than 40 companies had deployed it.

That matters because Glean's public launch was not the start of customer discovery.

The company had already spent roughly two years building with real companies.

The design-partner strategy did three things at once.

It helped Glean improve the product.

It gave Glean proof that employees actually used it.

And it created references that made selling to the next customer easier.

This was Glean's first growth loop:

better customers produced more usage data → usage improved the product → strong usage produced customer references → references helped win more customers.

For enterprise software, that can be more valuable than trying to generate thousands of top-of-funnel signups.

The hard part was connecting all of a company's data

A simple search bar hid an ugly technical problem.

Every company stored information across different applications.

Every application had a different API and data model.

And company data had permissions.

An employee might be allowed to read one Google Doc but not another. A Salesforce record could be restricted by role. A Slack channel could be private.

Glean had to bring data from all these systems into one index without showing employees information they were not allowed to see.

That required building connectors that understood content, metadata, identities, permissions, updates, and deletions.

Then Glean had to rank all that information.

The internet gives Google enormous amounts of behavior data. Millions of people can search the same phrase and teach Google which result is useful.

A company might have only a few hundred or thousand employees.

Glean needed different signals.

It built what it calls an Enterprise Graph: a model of the relationships between employees, teams, documents, activity, and company knowledge.

That could help Glean understand that an engineering document was more useful to an engineer, or that one document was more authoritative and current than another.

This work took years.

It also became one of the biggest reasons Glean was ready when generative AI arrived.

Search was the wedge that got Glean inside companies

Connecting to a company's internal systems requires trust.

A CIO or security team is unlikely to approve broad access to company data for a product with no obvious value.

Search gave Glean a reason.

Employees immediately understood why they might want one search box for all their work.

Once a company completed the security review and connected its systems, Glean had something valuable: a permission-aware index of company knowledge.

The initial product solved search.

The underlying infrastructure could eventually power much more.

This became the foundation for Glean's later AI assistant and agents.

Glean sold through CIOs instead of relying on self-serve growth

Glean looks simple from the employee's side.

Open a search box. Type a question. Find an answer.

The buying process is very different.

An individual employee cannot reasonably enter a credit card and give a startup permission to crawl the company's documents, support tickets, source code, messages, and customer data.

Glean therefore built a traditional enterprise sales motion.

"We typically go and work with the CIO or the technology team at our customer," Jain explained in an Unusual Ventures interview.

That produced a useful combination.

The purchase happened from the top down.

Adoption happened from the bottom up.

IT could approve and deploy Glean centrally. Employees then decided whether the product was useful enough to become part of their daily work.

Strong employee usage made it easier to expand.

Confluent later said more than 70% of employees became active Glean users. Glean's customer case study says Confluent attributed 15,000 hours of monthly time savings to the product.

Duolingo attributed more than 500 hours of monthly savings and $1.1 million in annual time value to Glean.

Those numbers come from Glean customer stories, so they should be treated as vendor-reported results rather than independent benchmarks.

Still, they show how Glean sold the product internally: save employees time and make company information easier to use.

ChatGPT changed Glean's growth curve

Glean's biggest growth inflection came after November 2022.

Before ChatGPT, Glean already used machine learning and language models for semantic search.

But the main user experience was still search.

You asked for something.

Glean returned relevant documents and links.

ChatGPT showed people a new interface.

Instead of finding the document containing an answer, users could ask a question and receive the answer directly.

For Glean, this was a major opportunity.

The hard part of an enterprise AI assistant is not only generating text.

The assistant also needs access to the company's private information. It needs to know which information is relevant. It needs to respect permissions. And it needs to keep the answer grounded in sources.

Glean had already spent years building those pieces.

Soon after ChatGPT launched, the company moved quickly to build a conversational assistant on top of its search infrastructure.

The result eventually became Glean's AI assistant.

The product could retrieve relevant company information and give it to a large language model, which could then answer the employee's question.

Glean went from helping people find documents to helping them get answers.

Demand changed too.

Jain later described the difference clearly: before ChatGPT, Glean had to create the enterprise-search market. After ChatGPT, companies were actively looking for ways to use generative AI with their internal data.

Forbes reported that Glean had more than 200 enterprise customers by November 2023.

ChatGPT did not create Glean's infrastructure.

It made that infrastructure much more valuable.

Glean increased its sales capacity as AI demand exploded

A hot market does not automatically become revenue.

Glean also changed how aggressively it sold.

According to an account from SoftBank Vision Fund, Glean doubled its workforce hiring target during 2023 and tripled the size of its sales team over roughly a year.

That was a logical response to the shift in demand.

Before ChatGPT, Glean had to persuade companies that enterprise search mattered.

After ChatGPT, many CIOs already wanted enterprise AI.

Glean could now enter the conversation with several years of enterprise-search infrastructure already built.

Revenue began rising quickly.

CNBC reported that Glean had about $39 million in annualized revenue in January 2024, up from roughly $10 million a year earlier.

In February 2024, Glean raised more than $200 million at a $2.2 billion valuation.

By September, it raised another $260 million at a $4.6 billion valuation.

The company said annual recurring revenue had more than tripled over the previous year.

Glean expanded from search to an AI platform

Search got Glean into the company.

The next step was increasing the number of things customers could do once Glean was connected.

Glean added conversational answers.

Then it added ways for companies to build custom AI applications.

Then it moved into agents.

This is a common way enterprise software expands.

The initial product solves one clear problem.

Once the vendor has data access, users, trust, and distribution, it can add adjacent products.

For Glean, the expansion looked like this:

Search → Assistant → Apps → Agents → autonomous work

The same underlying company context could power each product.

That increased Glean's potential contract size without requiring the company to acquire a completely new customer every time it launched something.

Glean crossed $100M ARR in early 2025

In February 2025, Glean announced that it had reached $100 million in annual recurring revenue during its fiscal fourth quarter.

It said its customer base had more than doubled during the previous year.

Glean also reported that the average user made about five queries per day.

Its weekday daily-active-to-monthly-active user ratio was around 40%.

These are company-reported metrics, but they explain why search can be a strong enterprise wedge.

Search is naturally frequent.

People forget where documents are.

They need answers every day.

A product that becomes the default place to ask those questions can create unusually high usage for enterprise software.

Habitual usage also makes expansion easier.

Agents gave Glean another product to sell

Glean launched Glean Agents in February 2025.

The idea was to move beyond finding and summarizing information.

An agent could use company context to complete work.

For example, it could gather information from several sources, produce a result, or take actions in connected systems.

The important part for Glean was that agents used the infrastructure it had already built.

An enterprise agent needs context.

It needs access to the right applications.

It needs permissions.

It needs to know which information the user can access.

It needs governance.

These were close to the same problems Glean had been solving since 2019.

Agents therefore expanded the market around Glean's existing product rather than forcing the company to start again.

Large customers became a bigger part of Glean's growth

Glean's next growth engine was account expansion.

In June 2025, the company raised $150 million at a $7.2 billion valuation.

By December 2025, Glean said it had reached $200 million ARR, nine months after passing $100 million.

The company said its number of company-wide deployments had more than doubled.

It also said the number of contracts worth at least $1 million had nearly tripled.

This suggests that growth was increasingly coming from larger deployments and bigger contracts, not only from adding new customers.

By May 2026, Glean reported another major milestone:

$300 million ARR.

That meant it had grown from $100 million to $300 million in about 15 months.

Glean also said its Fortune 500 customer count had almost doubled year over year.

More importantly, 85% of customers used Glean across at least five departments.

That shows how the expansion motion works.

Glean can land through a specific use case or team, then spread across engineering, sales, support, marketing, HR, finance, and other departments.

The more parts of the company that use Glean, the larger the account can become.

Partnerships helped Glean get through enterprise procurement

Glean also expanded distribution through cloud providers and other enterprise software companies.

It became available through marketplaces and partnerships involving AWS, Google Cloud, and Microsoft.

This matters in enterprise sales.

A customer may already have a large spending commitment with a cloud provider. Buying software through that provider's marketplace can make procurement easier.

Cloud partners can also help with co-selling, security, infrastructure, and access to large enterprise accounts.

Glean later added partnerships and integrations involving companies such as Databricks, Snowflake, Workday, Dell, and Palo Alto Networks.

These relationships come with a tradeoff.

Some of Glean's largest partners also control platforms that could compete with it.

Microsoft has Copilot.

Google has Gemini.

OpenAI and Anthropic sell directly to enterprises.

Glean therefore needs to remain useful even when companies have access to AI from their existing software vendors.

Its bet is that companies will continue using many different applications and models, creating a need for one layer that can understand information across all of them.

The growth engine behind Glean

Glean's growth can be reduced to a few reinforcing loops.

The strongest part of this model is that the pieces reinforce one another.

More integrations make the product useful for more companies.

More company data can make search and AI answers more useful.

More useful answers increase employee usage.

Higher usage helps justify wider deployments.

Wider deployments increase contract size.

And every new AI use case gives Glean another reason to use the same underlying company data.

What made Glean hard to copy?

Glean's advantage was not simply adding an LLM to company documents.

A new startup could build an impressive AI chat demo quickly.

The harder work was everything behind the chat box.

Glean had spent years building connectors.

It had to understand identities and permissions.

It had to keep indexes updated.

It had to rank information from different systems.

It had to learn how people, teams, and documents inside each company were related.

It had to pass enterprise security reviews.

And it had to build relationships with CIOs.

By the time generative AI became a major enterprise priority, Glean already had this foundation.

That gave it a head start.

Why ChatGPT helped Glean instead of killing it

ChatGPT could have looked like a threat.

If large language models could answer questions, why would companies still need enterprise search?

The answer was context.

A public model did not automatically know a company's product plans, customer history, internal documentation, policies, Jira tickets, Slack conversations, or Salesforce records.

Giving a model useful company knowledge required retrieval.

It also required permissions and governance.

Those were problems Glean already knew how to solve.

So the rise of ChatGPT increased the value of Glean's search infrastructure.

The interface changed.

The underlying need for good retrieval became more important.

Glean's reported ARR needs one caveat

Glean reported reaching $300 million in ARR in May 2026.

The company said this was three times the $100 million it had reported 15 months earlier.

However, Glean's pricing had expanded beyond simple fixed subscription contracts.

Jain told TechCrunch that customers could use consumption pricing or hybrid plans that combined an active-user fee with model usage charges.

Usage-based revenue can change from month to month.

That means part of Glean's reported $300 million figure is better understood as annualized revenue run rate rather than fixed subscription ARR.

Glean is also private.

It does not publish audited public financial statements showing recognized revenue, customer concentration, or net revenue retention.

The $7.2 billion valuation is also a private financing valuation.

It came from Glean's June 2025 Series F.

It should not be confused with revenue or the market capitalization of a public company.

As of September 12, 2026, $300 million remained Glean's latest publicly reported ARR figure, while $7.2 billion remained its latest disclosed financing valuation.

What founders can learn from Glean's growth

Glean's growth looks fast when you start the story in 2023.

It looks very different when you start in 2019.

The company spent years on infrastructure before generative AI made the market explode.

That leads to the most useful lesson from Glean.

A startup cannot predict exactly when a platform shift will happen.

It can put itself in a position to benefit when one does.

Glean had already found a painful problem.

It had customers.

It had enterprise distribution.

It had a permission-aware data layer.

It had years of search technology.

When ChatGPT changed how companies thought about AI, Glean could move quickly because the difficult foundation was already there.

The AI wave accelerated Glean.

The work before the wave made that acceleration possible.

Top-down enterprise sales is hard to copy early. Being the name buyers already know when your category suddenly matters is not. Okara's LinkedIn Agent keeps your point of view in front of buyers every week, and the GEO Agent shows whether AI assistants mention you when those buyers start researching.

More startup growth stories

• Harvey AI Revenue and Valuation: How It Grew to $15.5B
• Sierra AI Revenue: How It Grew to $200M ARR in 9 Quarters
• Perplexity Revenue: How It Grew to $750M and 100M+ Users
• All startup growth case studies

Frequently asked questions

How did Glean grow?

Glean grew by starting with enterprise search for fast-growing technology companies, working with free design partners, selling through CIOs, and expanding usage across each customer. After ChatGPT launched, Glean used the search, connectors, permissions, and company context it had already built to launch an AI assistant and later AI agents. Its growth accelerated as companies increased spending on enterprise AI.

What was Glean's initial growth strategy?

Glean's early growth strategy was based on design partners rather than self-serve product-led growth. The company gave its product to about 20 companies for free, studied how employees used it, improved the product, and converted early deployments into paid enterprise contracts.

Who founded Glean?

Glean was founded in 2019 by Arvind Jain, T.R. Vishwanath, Piyush Prahladka, and Tony Gentilcore. Jain previously spent more than a decade at Google and co-founded Rubrik.

What problem does Glean solve?

Glean helps employees find and use information spread across workplace applications such as Google Drive, Slack, Salesforce, Jira, GitHub, and other enterprise systems. It later expanded from search into AI assistants and agents that use the same company context.

Is Glean product-led?

Glean has strong product usage, but its sales model is primarily enterprise-led rather than classic self-serve PLG. Glean usually works with CIOs, IT teams, and security teams because deploying the product requires access to company systems and data.

How did ChatGPT affect Glean?

ChatGPT changed the way companies thought about enterprise search. Glean could use its existing search infrastructure to retrieve private company information and let large language models turn that information into direct answers. Demand for enterprise AI also increased sharply after ChatGPT's release.

What is Glean's ARR?

Glean reported reaching $300 million in annual recurring revenue in May 2026, up from $100 million 15 months earlier. Because Glean also uses consumption and hybrid pricing, part of that number may be closer to annualized usage run rate than traditional fixed subscription ARR.

What is Glean's valuation?

Glean's latest disclosed financing valued the company at $7.2 billion. The valuation came from a $150 million Series F announced in June 2025. Glean remains a private company.

Why was enterprise search a good wedge for Glean?

Search solved an immediate problem employees understood. It also gave CIOs a reason to connect company applications and complete the required security work. Once those systems were connected, Glean could use the same data and permissions layer to power AI assistants, applications, and agents.

What is Glean's main competitive advantage?

Glean's main advantage is the infrastructure it has built around enterprise data: connectors, indexing, permissions, search relevance, and its Enterprise Graph. This lets Glean give AI systems access to company information while respecting existing access controls.

Sources

• Finding what you need at work shouldn't be so hard — Arvind Jain
• How Glean found product-market fit — Unusual Ventures
• The evolution of enterprise AI — Madrona
• How Glean solved enterprise search — Sequoia
• Built two unicorns in 12 years — EO interview with Arvind Jain
• Glean emerges from stealth — Forbes
• Glean's $100 million Series C — Glean
• Glean's $200 million Series D — Reuters
• Glean's $260 million Series E — Reuters
• Glean's $150 million Series F — Reuters
• Glean reaches a reported $300 million ARR — TechCrunch
• Confluent customer story — Glean
• Duolingo customer story — Glean
• Databricks customer story — Glean
• Connector authentication requirements — Glean documentation
• Crawling and learning process — Glean documentation
• Crawler and indexing limits — Glean documentation
• Slack Real-Time Search migration — Glean documentation